Laurence found that people living in diverse and segregated areas had rather negative attitudes towards immigrants, while people who lived in diverse and integrated areas had positive attitudes.
But what do these areas look like?
Basic takeaway: Laurence’s results are not as straightforward to interpret as they seem, since it is unclear what exactly he is measuring.
So far, we have examined evidence demonstrating that increasing diversity leads to anti-minority attitudes and behaviors. But why does this relationship exist? A mechanism tell us “why”.
Consider Durkheim’s classic study of suicide, summarized from The
Practice of Social Research by Earl Babbie, pp.339-340.
Durkheim noticed two empirical patterns in suicide rates:
Durkheim thought there was a single mechanism linking these two patterns. In particular, he suggested that many suicides are a product of anomie or “normlessness”, or a general sense of social instability and disintegration. During times of political unrest, people feel that the old ways of society are collapsing, and suicide is one answer to this severe discomfort. In contrast, social integration and solidarity offer protection against suicide. And Catholicism, as a more structured and integrated religious system, gave people a greater sense of coherence and stability.
In this story, feelings of anomie are the mechanism which explains the internal logic behind the broader empirical patterns we observe.
Note however that Durkheim never directly tested this mechanism. Testing mechanisms is what we’ll focus on over the 2 weeks.
We know from Enos (and other experimental work) that exposure to immigration can cause whites to adopt anti-immigration attitudes. One influential set of mechanisms focuses on feelings of threat induced by immigration.
Your readings mentioned several different types of threats related to the economic and cultural impacts of immigration, as well as concerns about security and safety.
Social scientists also often talk about egocentric vs sociotropic concerns over immigration.
Small groups: in which cells of the table would you place the following statements:
| A Typology of Threats: Reasons Why Natives May Dislike Immigrants | ||
| Egocentric | Sociotropic | |
|---|---|---|
| Economic Threat | ? | ? |
| Cultural Threat | ? | ? |
| Safety Threat | ? | ? |
Here’s a thought experiment from When Ways of Life Collide pp.80-81:
You join two neighbors in the middle of a conversation. One confides in you. Taking in immigrant minorities has provided a pool of cheap labor; already it has cost some of her friends their jobs; now she fears it threatens hers. The second neighbor joins in, agreeing that immigrants threaten a lot of people’s jobs and, what is more, immigrants suck up government assistance, costing Dutch society more and more each year, and have driven up the crime rate.
You have known both neighbors for a long time. Previously the first neighbor did not complain about immigrants — indeed, she had expressed some sympathy toward them and their problems. The second has complained about immigrant minorities for as long as you have known him…In fact, he has always disliked almost everything about minorities — their talking a foreign language in public, their offensive behavior, the odd clothes they wear, and so on and so forth.
Both neighbors are critical of immigrant minorities. But they differ in a key respect. One perceives that immigrants pose an economic threat; the other dislikes them pure and simple.
Imagine researchers conduct a survey where they ask two questions:
They find a strong correlation: people who answer that immigrants threaten “native” jobs are also much more likely to support immigration restrictions.
Discussion: Imagine your friend reads about this study in the newspaper, and says: “Oh, that’s clear evidence that economic threats explain anti-immigrant attitudes!” What do you think about this interpretation?
Suppose you wanted to test the mechanism that cultural threat drives attitudes towards refugees. Your hypothesis is that refugees from majority-Muslim countries (e.g. Syria) are perceived as more culturally threatening – and are thus more disfavored – than refugees from majority-Christian countries (e.g. Ukraine).
You recruit a representative sample of native Germans to participate in a survey. You randomly divide survey respondents into two groups.
Both groups receive the following prompt:
“Imagine you live in a small town of around 1000 people. Last year, because of reduced tourism due to the COVID pandemic, your town’s hotel was forced to close. Recently, you learn that the empty hotel building will be converted into refugee accommodation for 30 asylum seekers from _____.”
For group A _____ is filled in with Syria, while for group B _____is filled in with Ukraine.
Respondents in both groups are then asked how much they support or oppose the hotel conversion.
Discussion: By comparing the average reponses of group A vs B, do you think you have isolated the effect of cultural threat on attitudes towards refugees? Or, put another way, what are some other (non-cultural) differences between Ukrainian and Syrian refugees that may also influence attitudes towards them?
Here, the problem is that Ukrainian and Syrian refugees differ not only
on the religious dimension, but also on dimensions such as gender or
education. So we have what is called a compound
treatment. In this case, while we have identified a treatment
effect, we don’t know what element of the compound treatment is “doing
the work.”
“Imagine you live in a small town of around 1000 people. Last year,
because of reduced tourism due to the COVID pandemic, your town’s hotel
was forced to close. Recently, you learn that the empty hotel building
will be converted into refugee accommodation for 30 asylum seekers from
_____. You have further been informed that the refugees will all be
women and children who are fleeing violence in their home countries.
These refugees have already been in Germany for several months, and have
learned basic German skills through prior participation in organized
integration courses.”
Your task is to design a scientific study which would test one or more
of the mechanisms we discussed. Imagine you had the opportunity to add a
survey experiment to an existing high-quality representative survey in
Germany. Please discuss:
Write up your experimental prompt in Padlet and be prepared to present it to the class.
Social Identity Theory
The above mechanisms posit that prejudices arise from group conflicts over economic and symbolic (cultural) resources.
But what if the simple act of categorizing people into groups gives rise to bias and in-group favoritism?
This is the main insight of Social Identity Theory.
Of course, real-life groups come with a lot of “baggage” – tradition, shared experience, socialization, common institutions, etc. The trick is in demonstrating the causal effect of categorization, free from all of this baggage. This was done through experiments using the minimum group paradigm.
Allocation to Groups
How many dots are in the this image?
Experimental participants are then divided into “overcounters” and “undercounters”.
Measuring Group Bias
Which allocation would you prefer?
Notice that Option A maximizes absolute gains for the ingroup, but minimizes the ingroup-outgroup “gap”. Alternatively, picking Option B makes your group relatively better off, but absolutely worse off.
Another way of thinking about this: if you pick Option B, you are paying a cost to harm the outgroup.
Food for thought:
If the mere act of categorization leads to bias, what do you think about the following prejudice-reduction strategies?
Implications of Conflict Theory vs. SIT for intergroup relations: